@vishalmisra As you correctly point out “A proof is a chain of steps you are sure of”
Then the problem is how to search over these chains, or representations of these. I am not sure of searching over embedding space. There can be other ways.
2/ One possibility: the bottleneck isn’t search.
The hard part is finding the representation where the proof becomes a short walk. Execution scales. Finding the low-entropy structure may not.
More here: https://t.co/Y4oYGGlUyY
This is why we need to gate-keep science. Two pseudo-intellectuals thinking they discovered something deep, conflating
>social attention
>transformers attention
>quantum physics observer (attention)
These have nothing in common, other than the ambiguity of English language. Naked ladies on Instagram have nothing to do with a weighted average followed by softmax.
But they're both so mind-blown by their discovery. Dunning–Kruger will only get amplified by AI sycophancy.
Please call me out if you see me going beyond my own DK threshold.
@Cockroach4India Let’s analyse from a system design perspective.
You gotta understand that your tech stack is dependent on other corporations and your operating principles are not decentralised.
You have many single points of failures along the way.